A shop structure generator for cell formation research

Publisher Website
Google Scholar
Abstract
We present a new computerized shop generator designed for research related to cell formation. The generator can take advantage of the following input data: number of parts and machine types in the population, part-machine matrix density, part and machine categories and proportions of parts and machines belonging to each category (the part categories classify parts depending on the number of machine types that appear in their routeings; similarly, the machine categories classify machines depending on the fraction of the part population they process), probabilities to determine machine sequence patterns (including possible revisit-ations), part demand, unit processing times, productive capacity per machine type, and maximum allowable utilization per machine. As part of the output, the shop generator creates randomized part-machine matrices. It can also calculate the required number of machines of each type and their utilizations, and produce machine sequences with and without backtracks. A unique feature of the generator is that it does not create part-machine data by reshuffling matrices with known block structures (i.e. the solution to the cell formation problem is unknown). Rather, data characteristics of the shop(s) to be cellularized are entered into the generator as input parameters. These parameters can be either hypothetical or based on typical characteristics found in industrial manufacturing data bases. The shop generator software, programmed in Turbo Pascal, and a user's manual are available from the second author.

This publication has 5 references indexed in Scilit: